ECE 750/SDE625 Final Project

$30.00

Category: You will Instantly receive a download link for .zip solution file upon Payment || To Order Original Work Click Custom Order?

Description

5/5 - (2 votes)

Instructions
The final report should be 10 to 12 pages length according the style and format outlined in this link. The
report should include (besides the name and affiliation) an abstract, an introduction, background of the
problem and literature review, fundamentals of the technique(s) proposed to solve the problems,
simulation (experimental results) and concluding remarks. The report should also include as many
references as possible. Please check any IEEE, ASME, or any other transactions or journals for help in
writing articles.
The main goal of this exercise is to have the student produce a report that looks as much as possible as a
journal article ready for submission.
The final reports are due on August 3, 2012 (before 11:30am) and should be uploaded on line in the
folder final with the title finalproject_yourlastname.doc and finalproject_yourlastname.pdf and
finalproject_yourlastname.zip (including all files and matlab files you used to run the programs). The
upload ftp site is located atftp://pami.uwaterloo.ca (username:ece750s10; password:750s10). Use a
dedicated ftp program such as the ftp explorer (rather than using IE or chrome web browser ). Note that
files you upload should be in the correct directory. Files that you upload will be hidden as soon as you
upload them.
Final Project
We would like to investigate how to use genetic algorithms for neural networks. The requirement from
students is to read some literature on the subject and then write up to 12 pages report (in the same
format as the one for the paper analysis). The report needs to include the following:
 Abstract
 Introduction
 Various aspects of using GA for Neural Networks (literature review)
 GA for optimization of Neural Networks (as an alternative to gradient descent)
 Simulation/Application for an example of your choice (either classification or function
approximation)
 Analysis of the Results (with the use or without use of GA)
 Conclusion
You may wish to use some of the link below:
http://www.cs.colostate.edu/~genitor/MiscPubs/gann.pdf
http://www.itba.edu.ar/nuevo/archivos/secciones/art_revistas_46.pdf
http://www.generation5.org/content/2000/nn_ga.asp